Synthetic human-like fakes: Difference between revisions
Juho Kunsola (talk | contribs) + In November 2024, Nvidia researchers announced they have made and trained a Foundational Generative Audio Transformer |
Juho Kunsola (talk | contribs) |
||
| Line 212: | Line 212: | ||
In November 2024, Nvidia researchers announced they have made and trained a [https://fugatto.github.io/ Foundational Generative Audio Transformer (Opus 1) at fugatto.github.io] or Fugatto for short. | In November 2024, Nvidia researchers announced they have made and trained a [https://fugatto.github.io/ Foundational Generative Audio Transformer (Opus 1) at fugatto.github.io] or Fugatto for short. | ||
The researchers state ''Fugatto is a versatile audio synthesis and transformation model capable of following | |||
free-form text instructions with optional audio inputs. ''<ref>https://research.nvidia.com/publication/2024-11_fugatto-1-foundational-generative-audio-transformer-opus-1</ref> | |||
=== Documented crimes with digital sound-alikes === | === Documented crimes with digital sound-alikes === | ||